A new time synchronization method for distributed simulations is proposed. In this approach, each federate estimates its potential synchronization point (PSP)). The coordinator, after receiving PSPs, selects the minimum of them (synchronization point (SP)). Each federate then advances its time to SP. While each federates advances its time, it re-calculates its PSP whenever the certain system status changes, and sends it to the coordinator. The coordinator updates SP when the updated PSPs affect it, and notifies its new SP to the federates. When a federate reaches close to the actual epoch event, the synchronization method switches to the conservative phase.